Title

COMMUNICATION ON LIMITED-MOBILITY UNDERWATER SENSOR NETWORKS

Introduction

Over 70% of the earth’s surface is covered in water. Sensing these bodies of water is a huge challenge to the scientific community and underwater sensor networks can help to overcome this challenge. An underwater sensor network is a collection of sensor nodes that collaboratively sense a body of water. In order to work together, the sensor nodes must have the capability to communicate with each other. The problem in question is the implementation of network communication between limited-mobility underwater sensor networks.

Purpose

In the field of underwater sensor networks it is important to figure out how the sensor nodes will communicate with each other. The network must be robust enough to handle packet-loss, the loss of data while communicating and have a communication protocol. The sensor nodes that the simulator is modeled after are capable of rising and descending in the water. This allows the sensor node to take advantage of two types of communication, radio when the node is surfaced and acoustic when the node is underwater.

Method

In order to better understand and solve our problem we built a custom simulator in MATLAB. The simulator uses important inputs such as network size, node locations, packet-loss rate, power usage parameters, communication range, etc. and is then able to simulate communication in the network. The simulator has 8 different message routing algorithms that utilize varying levels of system-knowledge. The network also has the capability of simulating packet-loss in acoustic communication. To test the effectiveness of the different algorithms and the effect of packet-loss we used network sizes of 10-100 nodes in 10 node increments and static topologies for each size. Acoustic packet-loss was tested at 0%, 1%, 10%, and 50%. The results are based on the average of multiple runs.

Results

Without packet-loss, the most power effective decentralized message routing algorithm was found to be the Greedy Shallowest Radio algorithm. The addition of packet-loss did not affect the ordering of the message routing algorithms.

Significance

We found that packet-loss has very little effect on the total power usage of the network. The total power usage of the network is dominated by the node movement and the node processing power. In terms of only the acoustic send power, there is a significant increase as the packet-loss rate is increased. We also found that near optimal message routing can be done with only the use of local node information. In future we plan to factor in optimal sensing coverage into the message routing algorithms and then implement the network architecture on actual hardware.

Location

DeRosa University Center, Stockton campus, University of the Pacific

Format

Poster Presentation

This document is currently not available here.

Share

COinS
 
Mar 25th, 10:00 AM Mar 25th, 3:00 PM

COMMUNICATION ON LIMITED-MOBILITY UNDERWATER SENSOR NETWORKS

DeRosa University Center, Stockton campus, University of the Pacific

Over 70% of the earth’s surface is covered in water. Sensing these bodies of water is a huge challenge to the scientific community and underwater sensor networks can help to overcome this challenge. An underwater sensor network is a collection of sensor nodes that collaboratively sense a body of water. In order to work together, the sensor nodes must have the capability to communicate with each other. The problem in question is the implementation of network communication between limited-mobility underwater sensor networks.